Mechanism for Carbachol-Induced Secretion of Lacritin in Cultured Monkey Lacrimal Acinar Cells

Abstract

Purpose.: Lacritin protein is highly expressed in the lacrimal gland, secreted into tear fluid, and detected only in primates. The mechanism for lacritin secretion has not been fully investigated, because a system for culturing primate lacrimal acinar cells had not been established. The purposes of the present study were (1) to develop a procedure to culture lacrimal acinar cells from monkey and (2) to determine the mechanism for the secretion of lacritin in the culture system.

Methods.: Acinar cells from monkey lacrimal gland were cultured and characterized. Lacritin and other proteins were detected by immunohistochemistry, immunocytochemistry, and immunoblot analysis. Secreted proteins were also detected in the medium from stimulated acinar cells. mRNAs were determined by microarray and qPCR. Intracellular calcium levels were measured by calcium-4 assay.

Results.: Acinar cells cultured for 1 day contained adequate amounts of lacritin, lactoferrin, and lipocalin for use in lacritin secretion studies. The cholinergic agonist carbachol (Cch) stimulated the secretion of lacritin and increased intracellular Ca²⁺. Cch-induced lacritin secretion was inhibited by the store-operated calcium (SOC) channel inhibitor YM58483 and the PKC inhibitors GF109203 and Ro-32-0432. Cch-induced lacritin secretion was not inhibited by MAPKK inhibitor U0126, although p42/p44 MAPK was phosphorylated. Cch also enhanced gene transcription, which was inhibited by U0126, GF109203, and calcium chelators.

Conclusions.: Successful culture of monkey lacrimal acinar cells showed that, among the prevalent tear proteins, the secretion of lacritin involved the PKC/Ca²⁺ pathway, not the p42/p44 MAPK pathway. Induction of transcription by Cch involved the independent p42/p44 MAPK and PKC pathways.

Lacrimal acinar cells synthesize and secrete proteins into the tear fluid. ^1,2 Well-known tear proteins include lysozyme, lactoferrin, lipocalin, and secretory IgA. ³ Lysozyme was decreased in blepharitis patients, ⁴ lactoferrin was lower in Sjögren and Stevens-Johnson syndromes, ⁵ and reduced lipocalin has been found in patients with seborrheic blepharitis and meibomian gland dysfunction. ⁶ Such observations have led to the concept that in addition to antibacterial function, certain tear proteins themselves can act as regulators of tear secretion, growth factors, or essential elements for renewal of ocular epithelia. 

Although less well studied, lacritin is the sixth most common mRNA in the National Eye Institute (NEI) human lacrimal EST database and codes for a 12.3-kDa protein. Lacritin is highly expressed in human lacrimal and meibomian glands. ⁷ Monkey lacritin has also been cloned. It shows 89% amino acid homology with human lacritin and is highly expressed in lacrimal gland and moderately expressed in conjunctiva and meibomian gland. ⁸ Lacritin orthologues have been reported in the Ensembl genome database (http://www.ensembl.org) in several species. ⁹ For example, the common shrew (Sorex araneus) and the cat (Felis catus) show 39% identity with human lacritin; however, to our knowledge, protein expression has not been reported. Lacritin promotes protein secretion from cultured rat acinar cells ¹⁰ and stimulates proliferation in cultured human corneal cells. ^11,12 Lower levels of lacritin have been observed in blepharitis patients. These preliminary data likewise suggest a role for lacritin in the maintenance of the ocular surface and raise speculation that topical lacritin could be used in the treatment of dry eye. ⁴  

Monkey models are important for proof of concept, but the mechanism for secretion of lacritin is not clear, because no culture system for primate acinar cells has been established. Thus, the purposes of the present study were (1) to develop a culture procedure for acinar cells from monkey lacrimal gland and (2) to determine the mechanism for lacritin secretion in cultured monkey acinar cells. 

Materials and Methods

Experimental Animals

Lacrimal glands from rhesus monkeys (Macaca mulatta, 1–16 years of age) were obtained at necroscopy from the Oregon National Primate Research Center (Beaverton, OR) from protocols not related to the present studies. Experimental animals were handled in accordance with the ARVO Statement for the use of Animals in Ophthalmic and Vision Research and the Guiding Principles in the Care and Use of Animals (DHEW Publication, NIH 80-23). 

Immunohistochemistry of Monkey Lacrimal Gland

Lacrimal glands were fixed in formaldehyde and embedded in paraffin. Three-micrometer sections were stained with H & E, or they were blocked with 1% bovine serum albumin (BSA) in phosphate-buffered saline (PBS) for 1 hour at room temperature. The sections were then incubated for 2 hours at room temperature with primary antibodies for lactoferrin (dilutions: 1:200; Sigma-Aldrich), lacritin (1:200), ⁸ lipocalin (1:200; Santa Cruz Biotechnology, Santa Cruz, CA), or vesicle-associated membrane protein 2 (VAMP2, 1:100; Assay Designs, Ann Arbor, MI). After the sections were rinsed with PBS, they were incubated with Alexa-Fluor 488-labeled anti-rabbit IgG (1:2000; Invitrogen Corp., Carlsbad, CA) or Alexa-Fluor 546-labeled anti-rabbit or anti-goat IgG (1:500; Invitrogen) for 1 hour at room temperature. Negative controls were treated in the same manner, but normal rabbit or goat IgG (1:200; Santa Cruz Biotechnology) were used in place of the primary antibody. Protocols for staining were the same as those for lacrimal gland tissue. Stained samples were photographed with an inverted microscope (Axiovert 200, equipped with an AxioCam MRc5; Carl Zeiss Vision, GmbH, Hallbergmoos, Germany). Images were compiled in image analysis software (Photoshop; Adobe, San Jose, CA). 

Acinar Cell Culture and Characterization

Lacrimal acinar cells were isolated, as described previously, ⁸ and cultured by using a modified protocol published by Hann et al. ¹³ Collagen I has been shown to be a suitable matrix for culturing rat acinar cells. ¹⁴ For this reason, monkey acinar cells in the present experiments were plated on collagen I (0.01 mg/cm²; BD Biosciences, Franklin Lakes, NJ)-coated plates with DMEM/Ham's F12 (Invitrogen) containing 10 ng/mL dexamethasone (Sigma-Aldrich.), 1 mM putrescine (Sigma-Aldrich), 50 ng/mL EGF (Invitrogen), 25 μg/mL l-ascorbic acid (Sigma-Aldrich), 1× ITS (Invitrogen), and 25 μg/mL gentamicin (Invitrogen). For proliferation assays, the cells were plated at 1 × 10⁵ cells/well in 24-well plates, and the number of cells was counted daily with a hemocytometer, up to 5 days. To confirm proliferation, the acinar cells were stained with Ki-67 (1:1000; Abcam, Cambridge, UK). Acinar cells were also stained with cytokeratin AE1/AE3 antibody (1:100; Millipore, Billerica, MA) to test for purity of the acinar cell cultures and absence of fibroblast contamination. For immunocytochemistry of tear proteins, the cells were plated at 2 × 10⁵ cells/well in 12-well plates, fixed with −20°C 100% methanol, incubated for 3 minutes at −20°C, rinsed with PBS, and permeabilized for 15 minutes with 0.1% Triton X-100 in PBS. Protocols for staining were the same as described for lacrimal gland tissue. 

For measurement of tear proteins, acinar cells were plated at 2 × 10⁵ cells/well in 12-well plates. At the end of the culture period, the cells were harvested into RIPA buffer containing 50 mM Tris-HCl (pH 7.4), 150 mM NaCl, 0.25% sodium deoxycholate, 0.5% NP-40, 1 mM EDTA, 1 mM phenylmethylsulfonyl fluoride, 1 mM sodium orthovanadate, protease inhibitor (complete Mini-EDTA free; Roche Diagnostics Corp., Indianapolis, IN), and phosphatase inhibitor cocktail I and II (EMD Biosciences Inc., San Diego, CA). The cells were then lysed by sonication, and protein concentrations were determined with a protein assay (Bio-Rad, Hercules, CA) with BSA standards. 

Measurement of Secreted Proteins

Cells were plated at 4 × 10⁵ cells/well in 12-well plates or 1 × 10⁶ cells/well in 6-well plates and cultured for 1 day. Protein secretion was induced by incubating the cells with carbachol (Cch; Sigma-Aldrich), PMA (EMD Biosciences) or ionomycin (EMD Biosciences) for 10 minutes at 37°C. The signaling pathways were tested by pretreating the cells for 30 minutes with the M1/M3 receptor antagonist atropine (Sigma-Aldrich), the nicotinic receptor antagonist mecamylamine (Sigma-Aldrich), the extracellular calcium chelator EGTA (Sigma-Aldrich), the intracellular calcium chelator BAPTA-AM (Invitrogen), the IP3 receptor antagonist 2-APB (Sigma-Aldrich), the SOC channel inhibitor YM58483 (Sigma-Aldrich), the MAPKK inhibitor U0126 (Sigma-Aldrich), and the PKC inhibitors GF109203 (EMD Biosciences) and Ro-32-0432 (EMD Biosciences). Detached cells were removed by centrifugation, and the culture medium was concentrated and subjected to immunoblot analysis. 

Immunoblot Analysis

Denatured protein samples were separated on 4% to 12% Bis-Tris gels (NuPAGE; Invitrogen) with MES buffer (Invitrogen). The proteins were then electrotransferred from the gels to polyvinylidene fluoride (PVDF) membranes (Millipore). The membranes were blocked with 5% skim milk in Tris-buffered saline with 0.05% Tween 20 (TTBS) and incubated overnight at 4°C with primary antibodies for lactoferrin (1:1000), lacritin (1:1000), lipocalin (1:100), GAPDH (1:1000; Sigma-Aldrich), p42/p44 MAPK (1:1000; Cell Signaling Technology, Inc., Danvers, MA), or phosphorylated p42/p44 MAPK (1:1000; Cell Signaling Technology) in 1% BSA/TTBS. The membranes were then rinsed in TTBS and incubated for 1 hour at room temperature with HRP-conjugated goat anti-rabbit secondary antibody (1:5000; Santa Cruz Biotechnology) or HRP-conjugated donkey anti-goat secondary antibody (1:5000; Santa Cruz Biotechnology). The protein bands were detected with chemiluminescence (ECL Plus; GE Health Care Corp., Piscataway, NJ), and images were captured (FluorChem FC2 imager; Alpha Innotech Corp., San Leandro, CA). 

Translocation of PKC and Phosphorylated p42/p44 MAPK

Subcellular fractionation was performed according to a published protocol. ¹⁵ The cells were sonicated in fractionation buffer containing 20 mM Tris-HCl (pH 7.5), 1 mM EDTA, 100 mM NaCl, 1 mM phenylmethylsulfonyl fluoride, 1 mM dithiothreitol, phosphatase inhibitor cocktail I and II, and protease inhibitor (Complete Mini-EDTA free; Roche, Indianapolis, IN). The cytosol fraction was collected after centrifugation at 16,000g for 40 minutes at 4°C. The pellets were rinsed twice with D-PBS, suspended in fractionation buffer containing 1% Triton X-100, and sonicated. After 30 minutes' incubation on ice, suspensions were centrifuged at 16,000g for 40 minutes at 4°C, to produce the supernatant membrane fraction. After measurement of protein concentrations with a protein assay (Bio-Rad), denatured proteins in LDS sample buffer were applied to 4% and 12% Bis-Tris gels (NuPAGE; Invitrogen) in MOPS buffer (Invitrogen). The protocol for immunoblot analysis was the same as just described, except anti-rabbit PKCα, -δ, and -ζ antibodies (Santa Cruz Biotechnology) were used at 1:200 dilution. To test the nuclear translocation of phospho-p42/p44 MAPK, the cells were induced for 5 minutes with Cch and stained for immunocytochemistry with a primary antibody against phospho-p42/p44 MAPK (1:1000). 

RNA Extraction and Reverse Transcription

Total RNA was extracted from cultured acinar cells (Trizol reagent; Invitrogen), according to the manufacturer's instructions. After phase separation, RNA was precipitated with 75% EtOH, bound to a fiberglass filter, washed, and collected with an RNA-extraction kit (RNAqueous; Ambion, Austin, TX). RNA quality was determined on a bioanalyzer (model 2100; Agilent Technologies, Palo Alto, CA). Total RNA was treated with DNase (DNA-free; Ambion) according to the manufacturer's protocol, with 1 U/μL RNase inhibitor (SUPERase-In; Ambion). The concentration of total RNA was determined (RiboGreen reagent; Invitrogen), and RNA was reverse transcribed at a concentration of 1 ng/μL with 10 U/μL reverse transcriptase (SuperScript II; Invitrogen), according to the manufacturer's instructions. For qPCR for NR4A1, IL-6, and PTGS2, the DNase reaction was not performed, because probes crossed exon boundaries. The optimized RT reaction conditions for the genes were 20 μM dNTP, 2 U/μL reverse transcriptase (SuperScript II; Invitrogen), and 0.5 U/μL RNase inhibitor (SUPERase-In; Ambion) in 1× first-strand buffer. 

Quantitative Real-Time PCR

Custom primers and probes for Rhesus macaque lacritin, lactoferrin, and lipocalin were developed by Applied Biosystems (ABI; Foster City, CA) by using gene expression assays (TaqMan; ABI; Table 1). Primers and probes for IL-6 (Rh02621719_u1) and PTGS2 (Rh01573476_m1) were purchased from ABI. NR4A1 and GAPDH FAM-BHQ-1 primers and probes were created (RealTimeDesign Software; Biosearch Technologies, Novato, CA) (Table 1). Primer and probe for 18s rRNA (cat. no. 4333760F) were obtained from ABI. 18s rRNA was used as a reference gene and normalization factor in qPCR for determining relative amounts of the transcripts for lacritin, lactoferrin, and lipocalin. GAPDH was used as the reference gene in the NR4A1, IL-6, and PTGS2 qPCR experiments. All assays except NR4A1 and GAPDH contained 250 nM probe, 900 nM each primer, 1× PCR master mix (ABI), and 1 ng cDNA. Reactions for NR4A1 were optimized at 300 nM forward/400 nM reverse primers and 250 nM probe. GAPDH was optimized at 100 nM forward/200 nM reverse primers and 250 nM probe. After an initial activation at 50°C for 2 minutes and 95°C for 10 minutes, PCR reactions were run for 45 cycles at 95°C for 15 seconds and 60°C for 1 minute. Fluorescence measurements were taken at every cycle after extension (Chromo 4; Bio-Rad). 

Table 1.

 

Primers and Probes for Quantitative PCR

Lacritin
Forward	5′-CCT CAA GCA GGC AGG AAC TA-3′
Reverse	5′-TCT TGC AAA GGC TTG TTC TGT TAG T-3′
Probe	5′-CCC CCT GAA ATC CA-3′
Lactoferrin
Forward	5′-GGT CCA CCT GAG GCC ATT G-3′
Reverse	5′-ACA CAG CTG GCT GAG AAG AAC-3′
Probe	5′-CTG GCC ACA GCT GCC T-3′
Lipocalin
Forward	5′-CCA CCT CCT GGC CTC AGA-3′
Reverse	5′-CGT CAT GGC CTT CAG ATA CCA-3′
Probe	5′-CCT GAC ACA TCC TGA ATC T-3′
NR4A1
Forward	5′-GCA CTG CCA AAC TGG ACT AC-3′
Reverse	5′-CCC AGC ATC TTC CTT CCC AAA G-3′
Probe	5′-TCC AAG TTT CAG GAG CTG GTG C-3′
GAPDH
Forward	5′-TGC ACC ACC AAC TGC TTA-3′
Reverse	5′-CAT GAG TCC TTC CAC GAT ACC AA-3′
Probe	5′-CCC TGG CCA AGG TCA TCC ATG A-3′

Quantification of gene transcripts in acinar culture samples was determined using the Pfaffl method ¹⁶ for normalization to 18s rRNA or GAPDH reference genes. PCR efficiencies for genes were determined to be similar and comparable, and the limit of detection for each primer and probe pair was determined to be well under the reported expression amounts in each sample. For lacritin, lactoferrin, and lipocalin, normalized expression levels were converted to the reported absolute copy numbers relative to a standard curve of known serial dilutions of a gel-purified and quantified PCR product (Quant-iT DNA HS kit; Invitrogen). 

Intracellular Calcium Measurements

Acinar cells were seeded at a density of 3 × 10⁴ cells/well in 96-well plates and cultured for 24 hours. The cells were then incubated at 37°C for 1 hour with the calcium-binding fluorophore calcium-4 (Molecular Devices; Sunnyvale, CA) in loading buffer (Hanks' balanced salt solution with 20 mM HEPES [pH 7.4]). After 20 seconds of initial background fluorescent measurements, Cch was added. Fluorescence measurements were performed on a microplate reader (Flexstation; Molecular Devices) every 1.75 seconds for 3 minutes. Some samples were pretreated with EGTA for 10 minutes or BAPTA-AM, 2-APB, or YM58483 for 30 minutes. Results are expressed as the ratio of maximum fluorescence to the average baseline fluorescence in each sample (F _max/F ₀). 

Gene Chip Analysis

Samples for microarray analysis were prepared from acinar cells incubated with or without 30 μM Cch for 1 hour. RNA extraction was performed as described earlier. Subjects consisted of replicate monkeys (7 and 9 years) with a normal and Cch-incubated sample from each animal. RNA was checked for quality (2100 Bioanalyzer; Agilent) and submitted for amplification, array hybridization, and detection (Affymetrix Microarray Core at the OHSU Gene Microarray Shared Resource Facility). Samples were hybridized to Rhesus macaque genome arrays (Affymetrix; Santa Clara, CA). Array data were normalized using the RMA algorithm and analyzed by ANOVA (Partek Genomics Suite; Partek Inc., St. Louis, MO). Signal, P-value, and detection calls for specific probes were determined by using the MAS 5.0 algorithm (Affymetrix), and PKC isotype probes were selected by sequences that align to the 3′ end of target mRNA and are grade A annotations. 

Statistical Analysis

Data were analyzed by Student's t-test or Dunnett's test. P < 0.05 was considered statistically significant (JMP software; JMP, Tokyo, Japan). 

Results

Establishment and Characterization of a Culture Protocol for Acinar Cells from Monkey Lacrimal Gland

H&E–stained sections of monkey lacrimal gland showed acini and ducts (Fig. 1A). These structures were similar to those observed in human lacrimal gland. ¹⁷ Immunohistochemistry visualized the major tear proteins lacritin, lactoferrin, and lipocalin in multilobed structures around the lumens of the acini (Fig. 1B). The vesicle marker VAMP2 revealed that lacritin was located in the secretory vesicles (Fig. 1B). ¹⁸ Lacritin partially co-localized with lactoferrin, but not with lipocalin (Fig. 1B). Lactoferrin partially co-localized with lipocalin. Lack of Ki-67 staining revealed that proliferation did not occur (Fig. 1B). 

Figure 1.

 

The presence of lacritin in acini from monkey lacrimal gland. (A) H & E staining of 1-year-old monkey lacrimal gland showing acini (black arrow), ductal structures (arrowhead), and lumens (white arrows). (B) Immunohistochemistry showing the presence of proteins for lacritin, lactoferrin, and lipocalin with different localizations. Lacritin co-localized with the vesicle marker VAMP2. No visible staining with the proliferation marker Ki-67 was observed. Rabbit and goat IgG were negative controls. White arrows: lumens. Scale bar, 20 μm.

Acinar cells cultured for 1 day attached to plates coated with collagen I (Fig. 2A). After spreading for 3 days, the cultured cells proliferated to a flattened morphology and subsequently became confluent on day 5. 

Figure 2.

 

Characterization of cultured acinar cells from monkey lacrimal gland. (A) Phase contrast microscopy showing growth of acinar cells in culture. Staining with Ki-67 (green) confirmed the proliferation of cultured cells at 3 days. Lacritin, lactoferrin, and lipocalin (red) were observed up to 3 days. Staining with cytokeratin (green) confirmed pure cultures of acinar cells. Lacritin co-localized with vesicle marker VAMP2 (red). Rabbit and goat IgG were negative controls. Scale bar, 20 μm. (B) Proliferation of cultured acinar cells from different monkeys at ages 7, 9, and 11 years. Data are the mean ± SD (n = 3). (C) Immunoblot analysis showed that lacritin, lactoferrin, and lipocalin in acinar cells decreased with culture time, whereas the control protein GAPDH was stable. (D) qPCR analysis showed decreased expression of mRNAs for three tear proteins. Data are the mean ± SD (n = 3).

The growth curves for acinar cells were similar among monkeys of different ages (Fig. 2B). Approximately 50% of acinar cells were attached to the plate 1 day after inoculation. Cell proliferation started on day 2 and continued gradually for 5 days. Cells staining positive with the proliferation marker Ki-67 appeared at day 3 and disappeared at day 5, suggesting proliferation and then contact inhibition (Fig. 2A). All cultured acinar cells stained with the epithelial cell marker cytokeratin during the culture period, suggesting pure cultures of acinar cells (Fig. 2A). 

Lacritin, lactoferrin, and lipocalin were detected by immunostaining in cultured acinar cells on day 1, but the staining decreased with culture time (Fig 2A). Immunoblot analysis performed on acinar cells at each day of cultivation confirmed an inverse relationship between the relative amounts of these tear proteins and acinar cell proliferation (Fig. 2C). Lacritin slightly decreased in cells cultured for 1 day. After 2 days, lacritin was significantly decreased and nearly depleted by 5 days. Lactoferrin and lipocalin were similarly decreased. The ubiquitous protein GAPDH was stable over the same culture period, suggesting that the cells become less differentiated as culture time increases. 

Expression of mRNA for lacritin was higher than that of lactoferrin and lipocalin (Fig. 2D). mRNAs for all three proteins decreased slightly on day 1 of culture and then showed major decreases at day 2. These results followed their specific protein levels, suggesting that in addition to constitutive secretion in cultured cells, control of tear protein secretion is partially regulated by transcriptional regulation. 

In view of these results, cells cultured for 1 day seemed most appropriate for studies on acinar secretion. Acinar cells cultured for 1 day showed epithelial morphology and granular regions in the cytoplasm (Fig. 3). Cellular distribution and co-localization of each tear protein were distinct in acinar cells cultured for 1 day. Lacritin showed patchy distribution throughout the cytoplasm, lipocalin showed punctuate staining in the cytoplasm, and lactoferrin showed grid-like staining with connections between patches. No co-localization between lacritin and lipocalin was observed in cultured cells. However, both proteins partially co-localized with lactoferrin (merged images in yellow). These results in cultured cells were similar to that of lacrimal gland (Fig. 1B), suggesting that acinar cells at 1 day of culture retained their differentiated character as observed in vivo. 

Figure 3.

 

Phase-contrast microscopy and immunohistochemistry in acinar cells cultured for 1 day, showing the presence of lacritin, lactoferrin, and lipocalin with differing localization. Scale bar, 20 μm.

Secretion of Lacritin from Cultured Acinar Cells

The endogenous cholinergic agonist Ach induced secretion of lacritin from acinar cells in a dose-dependent manner, with a minimum effective concentration of 1 μM (Fig. 4A). Secretion of lactoferrin and lipocalin was also induced by Ach at similar concentrations, although different cellular distributions of lacritin, lactoferrin, and lipocalin were observed in the cultured cells (Fig. 3). Similar to Ach, the acetylcholine analogue Cch induced secretion of lacritin, lactoferrin, and lipocalin (Fig. 4A). 

Figure 4.

 

Secretion of lacritin from cultured acinar cells. (A) Immunoblot analysis showing a dose-dependent secretion of lacritin, lactoferrin, and lipocalin by cholinergic agonists Ach or Cch. (B) Pretreatment with the muscarinic receptor antagonist atropine, but not the nicotinic receptor antagonist mecamylamine, completely inhibited secretion of tear proteins induced by Cch. The nicotinic receptor agonist nicotine did not induce secretion of proteins. Data are representative of results in three experiments in cultured acinar cells from three different monkeys.

To determine which subtypes of cholinergic receptors were responsible for Cch-induced secretion of lacritin, cells were pretreated before Cch induction with atropine, a muscarinic antagonist of M1 and M3 receptors, or with mecamylamine, an antagonist of nicotinic receptors. Atropine, but not mecamylamine, showed inhibition of Cch-induced secretion of the three tear proteins (Fig. 4B). The nicotinic receptor agonist nicotine did not induce protein secretion. These results suggest that the muscarinic receptors were involved in protein secretion induced by Cch. Of the five muscarinic receptor subtypes, only mRNA for the M3 receptor was significantly expressed in cultured monkey lacrimal acinar cells, as determined by microarray analysis (Table 2). The M3 receptor was therefore likely to be involved in Cch-induced secretion of tear proteins, such as lacritin. 

Table 2.

 

mRNAs for Muscarinic Receptor Isotypes in Monkey Lacrimal Acinar Cells

Probe Set ID	Muscarinic Receptor	7 y		9 y		Detection Call
		Signal	P	Signal	P
Mmu.3886.1.S1_at	1	5.8	0.466	2.5	0.696	Absent
MmuSTS.1732.1.S1_at	2	5.1	0.837	4.7	0.697	Absent
MmugDNA.35530.1.S1_at	3	77.1	0.002	78	0.0005	Present
MmuSTS.804.1.S1_at	4	6.9	0.749	3.6	0.541	Absent
MmugDNA.24600.1.S1_at	5	6.9	0.679	12.9	0.660	Absent

Involvement of Intracellular Ca²⁺ in Cch-Induced Secretion of Lacritin

Free cytoplasmic Ca²⁺ was significantly increased in cultured acinar cells by Cch in a dose-dependent manner, starting at a minimum concentration of 3 μM (Fig. 5A). Since increased Ca²⁺ was associated with increased tear protein secretion (Fig. 4A), Ca²⁺ may be involved in the mechanism for protein secretion from acinar cells. Pretreatment with 100 μM of the IP3 receptor antagonist 2-APB completely inhibited the Cch-induced increase in intracellular Ca²⁺ (Fig. 5C) and protein secretion (Fig. 5B). This result suggests that IP3 is an intracellular messenger for Ca²⁺ mobilization in acinar cells treated with Cch. 

Figure 5.

 

Increased intracellular Ca²⁺ induced by Cch and inhibition by an IP3 receptor antagonist. (A) Free Ca²⁺ (F _max/F ₀ ratio) in the cytoplasm significantly increased in acinar cells after induction by Cch. Data are the mean ± SD (n = 3) at 3 minutes; *P < 0.05 relative to the control group (0 μM). (B) Pretreatment with 100 μM of the IP3 receptor antagonist 2-APB inhibited Cch-induced protein secretion at 10 minutes and (C) inhibited increased intracellular Ca²⁺ at 3 minutes. Data are the mean ± SD (n = 3); #P < 0.05 relative to group treated without Cch. *P < 0.05 relative to group treated with Cch.

Depletion of Ca²⁺ in the endoplasmic reticulum (ER) has been shown to trigger influx of extracellular Ca²⁺ via the SOC channel. ¹⁹ In our cultured acinar cells, Cch-induced secretion of lacritin, lactoferrin, and lipocalin was completely inhibited by 5 mM EGTA (an extracellular Ca²⁺ chelator) or 125 μM BAPTA-AM (an intracellular Ca²⁺chelator; Fig. 6A). These chelators reduced intracellular Ca²⁺ to normal concentrations (Figs. 6B, 6C). We reasoned that the amount of Ca²⁺ mobilized from the ER was not enough to trigger Cch-induced lacritin secretion and that influx of Ca²⁺ from the extracellular pool was necessary. To confirm this hypothesis, calcium ionophore ionomycin was tested and found to induce protein secretion in a dose-dependent manner (Fig. 6D). The effect of ionophore was inhibited by BAPTA-AM (Fig. 6D). 

Figure 6.

 

Involvement of Ca²⁺ in the secretion of acinar proteins. (A) Five mM EGTA or 125 μM BAPTA-AM inhibited Cch-induced protein secretion at 10 minutes. (B) EGTA (5 mM) or (C) BAPTA-AM (125 μM) significantly inhibited increased Ca²⁺ induced by Cch at 3 minutes. Data are the mean ± SD (n = 3). #P < 0.05 relative to group treated without Cch. *P < 0.05 relative to the group treated with Cch. (D) Ionomycin-induced protein secretion at 10 minutes and inhibition by pretreatment with 125 μM BAPTA-AM.

In Cch-treated cells, pretreatment with the SOC channel inhibitor YM58483 inhibited increased intracellular Ca²⁺ in a dose-dependent manner (Fig. 7B). YM58483 also inhibited lacritin secretion (Fig. 7A). These results suggested that the SOC channel was an entrance point for extracellular Ca²⁺. 

Figure 7.

 

Involvement of the SOC channel in influx of Ca²⁺. (A) Dose-dependent inhibition of Cch-induced protein secretion by the SOC channel inhibitor YM58483 at 10 minutes and (B) inhibition of increased Ca²⁺ at 3 minutes. Data are the mean ± SD (n = 3). #P < 0.05 relative to the group treated without Cch. *P < 0.05 relative to the group treated with Cch.

Involvement of PKC in Cch-Induced Secretion of Lacritin

Pretreatment with 10 μM of the PKC inhibitor GF109203 (an inhibitor of PKCα, -β, -δ, -ε, and -ζ) or Ro-32-0432 (an inhibitor of PKCα, -β, and -ε) reduced Cch-induced lacritin secretion in monkey acinar cells (Fig. 8A), suggesting that various PKC isotypes are involved in Cch-induced secretion of proteins. Monkey lacrimal acinar cells were therefore cultured with the direct PKC activator PMA. PMA induced protein secretion, which was inhibited by 10 μM of the PKC inhibitors GF109203 and Ro-32-0432 (Fig. 8B). The extracellular chelator EGTA (5 mM) did not inhibit PMA-induced secretion of lacritin, because PMA activates PKC directly. In contrast, the 125 μM intracellular chelator BAPTA-AM inhibited PMA-induced protein secretion (Fig. 8B). Thus, physiological levels of intracellular Ca²⁺ or Ca²⁺ mobilized from internal stores was necessary for PMA-induced secretion of proteins. 

Figure 8.

 

Involvement of PKC in protein secretion. (A) Application of 10 μM PKC inhibitor GF109203 or Ro-32-0432 inhibited protein secretion by Cch at 10 minutes. Data are representative of results in three experiments with acinar cells cultured from three different monkeys.(B) PMA (1 μM) induced secretion of lacritin, lactoferrin, and, lipocalin at 10 minutes. BAPTA-AM at 125 μM, GF109203 at 10 μM (GF), and Ro-32-0432 at 10 μM (Ro) inhibited secretion of proteins induced by PMA, but 5 mM EGTA did not. (C) Immunoblots for PKCα, -δ, and -ζ after induction by 1 μM PMA, showing increased PKCα only in the membrane fraction when measured at 0, 1, 5, and 10 minutes.

To further examine which specific isozymes of PKC were involved in secretion of tear proteins, we performed microarray analysis on acinar cells from monkeys of two ages (Table 3). Relative mRNA levels from PKC isozymes from a 7-year-old monkey based on signal strength, P-value and detection calls were ζ > δ > ν > ι > η > α > ε > μ > θ (column 4). Cells from a 9-year-old monkey gave similar results with minor differences (column 6). PKC protein expression of the classic (α), novel (δ), and atypical (ζ) subtypes was also demonstrated by immunoblot analysis (Fig. 8C). PKCα was primarily present in the cytoplasm, whereas PKCδ and -ζ were located in both membrane and cytoplasm areas (Fig. 8C). PKCα, but not PKCδ and -ζ, increased in the membrane fraction of cultured acinar cells 1 minute after PMA treatment. Since previous studies have shown that translocation of PKCα to the membrane causes activation, ²⁰ our data suggest that PKCα, needing Ca²⁺ for activation, is involved in PMA-induced protein secretion. 

Table 3.

 

mRNAs for PKC Isozymes in Monkey Lacrimal Acinar Cells

Gene Symbol	PKC	Probe Set ID	7 y		9 y		Detection Call
			Signal	P	Signal	P
PRKCZ	ζ	MmugDNA.40989.1.S1_at	511.9	0.0002	519.9	0.0002	Present
PRKCD	δ	MmuSTS.2318.1.S1_at	421.8	0.0008	358.1	0.0005	Present
PRKD3	ν	MmugDNA.40581.1.S1_at	343.4	0.0003	380.8	0.0002	Present
PRKCI	ι	MmugDNA.23670.1.S1_at	315.9	0.0002	201.8	0.0003	Present
PRKCH	η	MmugDNA.19823.1.S1_at	236.3	0.0002	302.9	0.0002	Present
PRKCA	α	MmugDNA.7366.1.S1_at	163.2	0.0002	281.8	0.0002	Present
PRKCE	ε	MmugDNA.10395.1.S1_at	83.7	0.0031	82	0.0026	Present
PRKD1	μ	MmuSTS.3032.1.S1_at	39.8	0.0193	46.4	0.0313	Present
PRKCQ	θ	MmugDNA.10942.1.S1_at	7.3	0.8371	3.9	0.7493	Absent
PRKCG	γ	MmugDNA.13748.1.S1_at	6.4	0.7805	4.7	0.8704	Absent
PRKCB	β1	MmugDNA.21588.1.S1_at	3.2	0.8286	2.3	0.8704	Absent

 Data are listed in decreasing levels of expression in 7-year-old animal.

Involvement of p42/p44 MAPK in Transcription, but Not in Protein Secretion, in Acinar Cells Cultured with Cch

Another kinase, p42/p44 MAPK has been shown to regulate protein secretion in rat lacrimal acini. ^21,22 When our monkey acinar cells were treated with Cch, p42/p44 MAPK was phosphorylated in a dose-dependent manner (Fig. 9A). Pretreatment with the 10 μM MAPKK inhibitor U0126 completely inhibited p42/p44 MAPK phosphorylation (Fig. 9B). U0126 did not affect protein secretion after induction by Cch (Fig. 9B) or PMA (Fig. 8B). Thus, unlike rat acinar cells, activation of p42/p44 MAPK by Cch was not important for Cch-induced protein secretion in cultured monkey lacrimal acinar cells. 

Figure 9.

 

Phosphorylation of p42/p44 MAPK was not involved in protein secretion. (A) Phosphorylation of p42/p44 MAPK was induced by Cch in a dose-dependent manner at 10 minutes. (B) The MAPKK inhibitor U0126 (10 μM) completely inhibited Cch-induced phosphorylation at 10 minutes but did not inhibit protein secretion. Data are representative of results three experiments in acinar cells cultured from three different monkeys.

Previous studies showed that activation of p42/p44 MAPK-regulated gene expression in a fibroblast cell line from Chinese hamster lung. ²³ The distribution of phosphorylated p42/p44 MAPK was therefore examined in our cultured monkey lacrimal acinar cells treated with Cch. Phosphorylated p42/p44 MAPK translocated into the nucleus after 5 minutes (Fig. 10A). Microarray analysis further showed that some mRNAs were significantly upregulated by Cch in acinar cells cultured from two different Rhesus macaques. These were the nuclear receptor family genes NR4A1, NR4A2, and NR4A3; the pleiotropic cytokines IL-6, LIF, and CLCF1; the CXC chemokine family genes CXCL1 and CXCL2; prostaglandin-endoperoxide synthase 2 (PTGS2), cysteine-serine-rich nuclear protein 1 (CSRNP1); inhibin βA (INHBA); salt-inducible kinase1 (SIK1), thyroid cancer protein 1 (TC-1); and cyclin-L1 (CCNL1; Table 4). Increased expression of NR4A1, IL-6, and PTGS2 was also confirmed by qPCR analysis (Fig. 10B), since these three genes had been reported to be transcribed by activation of p42/p44 MAPK. ^24–26 Transcription of the three genes by Cch was almost completely inhibited by BAPTA-AM, but phosphorylation of p42/p44 was not affected (Figs. 10B, 10C). This result suggests that Ca²⁺ may be necessary for transcription, but not for phosphorylation, of p42/p44. U0126 completely inhibited phosphorylation of p42/p44 and partially inhibited transcription of the three genes. GF109203 or EGTA partially inhibited transcription, but not phosphorylation, suggesting involvement of PKC in activation of transcription. Thus, transcription induced by Cch may be regulated by p42/p44 MAPK and PKC. 

Figure 10.

 

Involvement of p42/p44 MAPK and PKC in transcription. (A) Immunohistochemistry showing translocation of phosphorylated p42/p44 (red fluorescence, arrowheads) into the nucleus 5 minutes after treatment with Cch. (B) qPCR analysis after Cch-induction showing increased mRNAs for NR4A1, IL-6, and PTGS2 and inhibition by treatment with 125 μM BAPTA-AM, 10 μM U0126, 10 μM GF109203, or 5 mM EGTA at 1 hour. Data are the mean ± SD (n = 3). #P < 0.05 relative to the group without Cch. *P < 0.05 relative to the Cch-treated group. (C) Phosphorylation of p42/p44 MAPK induced by Cch was completely inhibited by the MAPKK inhibitor U0126 at 10 μM, but not by 10 μM GF109203, 5 mM EGTA, or 125 μM BAPTA-AM at 10 minutes. Data are representative of results in three experiments in acinar cells cultured from three different monkeys.

Table 4.

 

Upregulated Genes Induced by Cch in Monkey Lacrimal Acinar Cells

Gene Symbol	Gene Name	Probe Set ID	Change Ratio Cch vs. Nor	P
NR4A1	Nuclear receptor subfamily 4, group A, member 1	MmugDNA.18685.1.S1_at	74.60	0.038
NR4A3	Nuclear receptor subfamily 4, group A, member 3	MmugDNA.15738.1.S1_at	21.19	0.021
IL-6	Interleukin 6 (interferon, beta 2)	MmuSTS.4354.1.S1_at	19.23	0.019
IL-6	Interleukin 6 (interferon, beta 2)	Mmu.12240.1.S1_at	17.95	0.001
PTGS2	Prostaglandin-endoperoxide synthase 2	MmugDNA.4494.1.S1_at	16.75	0.002
LIF	Leukemia inhibitory factor (cholinergic differentiation factor)	MmugDNA.22098.1.S1_at	6.56	0.011
CXCL2	Chemokine (C-X-C motif) ligand 2	MmuSTS.1895.1.S1_at	4.04	0.007
INHBA	Inhibin, beta A	MmugDNA.21962.1.S1_at	3.77	0.003
NR4A2	Nuclear receptor subfamily 4, group A, member 2	MmugDNA.28332.1.S1_at	3.62	0.002
CSRNP1	Cysteine-serine-rich nuclear protein 1	MmuSTS.2136.1.S1_at	3.45	0.038
SIK1	Salt-inducible kinase 1	MmugDNA.1154.1.S1_at	3.44	0.017
TC-1	Thyroid cancer protein 1	MmugDNA.14772.1.S1_at	3.35	0.0004
CLCF1	Cardiotrophin-like cytokine factor 1	MmuSTS.3418.1.S1_at	3.11	0.016
CXCL1	Chemokine (C-X-C motif) ligand 1	MmugDNA.37457.1.S1_at	2.96	0.017
CCNL1	Cyclin-L1	MmugDNA.4307.1.S1_at	2.90	0.011

 Top 15 probes at P < 0.04 and change ratio >2. Nor, normal.

Discussion

The most important conclusions of the present study were that (1) monkey lacrimal acinar cells can be cultured and used to study pathways controlling tear protein secretion; (2) acinar cells cultured with our protocol respond to cholinergic agonists and secrete the tear proteins lacritin, lactoferrin, and lipocalin; (3) the pathway for Cch-induced tear protein secretion utilizes Ca²⁺ intracellular signaling and activation of PKC, but not the p42/p44 MAPK pathway; (4) Cch-induced transcription of NR4A1, IL-6, and PTGS2 involves both the PKC pathway and the p42/p44 MAPK pathway. 

Monkey Acinar Culture System

A culture protocol for lacrimal acinar cells from rat and rabbit has been reported. ^13,27 Our culturing system for monkey acinar cells responded similarly in some respects to these other animal cells, but is more relevant to the human situation. For example, the morphology of lacrimal gland from monkey was found to be similar to human (Fig. 1A), and the monkey has been used extensively as a model for human diseases and in drug trials because of the general similarities between monkeys and humans. 

Our cultured monkey acinar cells contained VAMP2-positive granules that stained positive for lacritin, lactoferrin, and lipocalin (Fig. 3). Such transport granules for tear proteins were also reported in rabbit. ²⁸ Our monkey acinar cells proliferated with increased culture time, but lacritin-positive cells decreased after day 1, delineating the need for using cells at 24 hours. Proliferation of acinar cells along with loss of differentiated characteristics with cultured time was also observed in rat and rabbit acinar cells. ^29,30  

Rabbit and rat acinar cells formed acinus-like structures when they were cultured on synthetic matrix (Matrigel; Invitrogen) for 2 to 3 days or 11 days, respectively, and they also secreted proteins such as β-hexosaminidase and peroxidase. ^18,29 Acinus-like structures were not morphologically detected in our cultured monkey cells. Differences in culture conditions, such as different extracellular matrices and culture periods, and/or different species may have a role in the formation of acinus-like structures. Our cultured cells at 1 day, however, did retain functional characteristics of acini observed in vivo, such as negligible proliferation and similar subcellular localization of tear proteins. Further, the monkey acinar cells cultured for 1 day responded to stimulators and secreted lacritin, providing a useful culture system for studying human-relevant, tear secretory mechanisms and their regulation. 

Cell Signaling Pathway for Lacritin Secretion

Previous studies in other animal models indicated an association between lacritin and tear protein secretion and cell growth. For example, lacritin enhanced tear secretion in rat lacrimal acinar cells ¹⁰ and stimulated corneal and salivary ductal cell proliferation. ^11,12 Lacritin also induced expression of membrane mucin MUC16 in human corneal epithelial cells (Laurie GW, et al. IOVS 2006;47:ARVO E-Abstract 1606). Cholinergic receptors have been observed in mouse lacrimal gland, ³¹ and cholinergic agonists induced an increase in cytoplasmic Ca²⁺ and activation of PKC, after secretion of tear proteins from lacrimal acinar cells from rats and rabbits. ^18,22  

In the present experiments with monkey acinar cells, Cch activated the muscarinic receptor M3, causing production of IP3 (Fig. 11). IP3 interacted with the IP3 receptor on the ER, causing Ca²⁺ mobilization from the ER, and induced Ca²⁺ influx into the cytosol via the SOC channel. Increased cytoplasmic Ca²⁺ activated the PKC necessary for protein secretion from the acinar cells. Preventing increased intracellular Ca²⁺ or activation of PKC inhibited protein secretion. 

Figure 11.

 

Hypothesized signaling pathway for Cch-induced transcription and regulation of lacritin secretion.

Our experiments did not provide direct evidence to show which specific PKC isozymes are responsible for tear protein secretion. Indirect evidence suggests that PKCα is likely to be one of the responsible isozymes. For example, of the classic isotypes, only PKCα was detected in our cultured acinar cells. Further, PMA caused translocation of PKCα to the membrane, a marker for activation of PKC. ²⁰ In previous studies in the rat, PKCα and -ε were both shown to be involved in Cch-induced secretion of proteins by lacrimal acinar cells. ^32–37 In the present monkey study, expression of PKCε was relatively low in microarray analysis, and protein was not detected by immunoblot analysis (data not shown). Further experiments are needed to determine whether other PKC isozymes are involved in the induction of monkey acinar cells. 

We also found that BAPTA-AM, but not EGTA, inhibited PMA-induced/PKC-activated protein secretion (Fig. 8). Our results suggest that Ca²⁺ from intracellular stores, and not influx from extracellular pools, is necessary for PMA-induced protein secretion. The source of increased cytosolic Ca²⁺ after PMA treatment has been reported to be cell-type dependent. ^38,39 All evidence taken together shows that increased Ca²⁺ and PKCα activation probably play important roles in the secretion of proteins from monkey lacrimal acinar cells. 

Tear proteins must be transported in secretory vesicles and fused with the plasma membrane for extracellular secretion. ⁴⁰ Indeed, the present experiments revealed that lacritin was located in the secretory vesicles (Fig. 1B). We speculate that cytoskeletal protein may be a key regulator for these final secretory steps. Other studies have shown that a dominant negative for PKCε inhibits remodeling of filamentous actin and decreases protein secretion in Cch-induced rabbit acinar cells. ²⁸ Ca²⁺ influx causes disassembly of filamentous actin under the plasma membrane, facilitating access of vesicles to proper sites for exocytosis in chromaffin cells. ^41–43 Thus, PKC and Ca²⁺ may induce protein secretion via modification of actin filaments in monkey acinar cells. 

Lack of a Role for the p42/p44 MAPK Pathway in the Lacritin Secretion Pathway

In our cultured monkey acinar cells, p42/p44 MAPK was phosphorylated after induction by Cch, and this phosphorylation was inhibited by U0126. Unexpectedly, this inhibition did not affect protein secretion. This finding was different from those in another study in rat lacrimal acini induced by Cch, in which p42/p44 MAPK was phosphorylated by Pyk2 and c-Src through Ca²⁺ and PKC. ²² The MAPKK inhibitor U0126 reduced phosphorylation of p42/p44 MAPK and enhanced protein secretion induced by Cch, suggesting negative regulation of protein secretion by activated p42/p44 MAPK. ²¹ Positive regulation was also observed by the finding that U0126 inhibits secretion of goblet cell glycoconjugates in rat conjunctival pieces induced by Cch. ⁴⁴ The role of p42/p44 MAPK appears to be diverse and dependent on the animal species and tissue. 

In the present experiments on monkey acinar cells, phosphorylated p42/p44 MAPK entered the nucleus after Cch induction (Fig. 10), suggesting the promotion of transcription. Tear proteins were candidates for such transcription, since production of new proteins for replenishment is necessary after secretion. mRNA expression for lacritin, lactoferrin, and lipocalin was not upregulated in response to Cch as determined by qPCR (data not shown), although the three genes already showed relatively high expression before induction. Signal intensities in microarray analysis of cells cultured from the seven-year-old monkey for lacritin, lactoferrin, and lipocalin were 18,391, 17,341, and 16,074, respectively (P < 0.01). These genes are most likely constitutively expressed at high levels, to maintain constant, continuously translated tear proteins. 

In contrast, of the 14 genes showing highest change ratios in microarray analysis after our monkey cells acinar cells were cultured with Cch, 7 have been described as being regulated through p42/p44 MAPK. ^{24–26,45–47} In our experiments with monkey, transcription of three of these genes: NR4A1, IL-6, and PTGS2, was inhibited by U0126, GF109203, EGTA, or BAPTA-AM (phosphorylation of p42/p44 MAPK was inhibited only by U0126). Transcription of these genes may also be regulated by the p42/p44 MAPK and PKC/Ca²⁺ in independent pathways in the monkey. Although not studied in our monkey experiments, the proteins from these genes are said in the literature to be essential in maintaining the ocular surface. NR4A1 is a known orphan nuclear hormone receptor for transcription ²⁴ and would help maintain acinar cells. IL-6 is a pleiotropic cytokine, promotes the survival and/or proliferation of primitive hematopoietic precursors, ⁴⁸ and may be important in the maintenance of corneal cells. PTGS2 leads to the generation of prostaglandins, which would protect mucosal ⁴⁹ and goblet cells. 

In conclusion, successful culture of monkey lacrimal acinar cells showed that, among the prevalent tear proteins, secretion of lacritin involved the PKC/Ca²⁺ pathway, and not the p42/p44 MAPK pathway. Both PKC and p42/p44 MAPK pathways, however, were involved in induction of transcription by Cch. This culture system provides a protocol that is more relevant to human studies on lacritin and for development of drugs to promote secretion of tear proteins in diseased conditions. 

Footnotes

 TRS is a paid consultant for Senju Pharmaceutical Co., Ltd., a company that may have a commercial interest in the results of this research and technology. MA and EN are employees of Senju Pharmaceutical Co., Ltd. These potential conflicts of interest were reviewed, and management plans approved by the OHSU Conflict of Interest in Research Committee were implemented.

Footnotes

 Disclosure: A. Morimoto-Tochigi, Senju Pharmaceutical Corp. (E); R.D. Walkup, Senju Pharmaceutical Corp. (E); E. Nakajima, Senju Pharmaceutical Corp. (E); T.R. Shearer, Senju Pharmaceutical Corp. (C); M. Azuma, Senju Pharmaceutical Corp. (E)

The authors thank Jennifer Rosales and Atsuko Fujii for technical assistance, Chris Harrington, PhD (Microarray Core Facility), for technical assistance in the DNA microarray assays and Cutting Edge Histologic Services, LLC (Tigard, OR) for histology services.